The so called ’spread-net’ of all those linked spreadsheets has created an unseen web of knowledge. A parallel computing universe, often undocumented and even unrecognised by the official corporate IT provider. This situation is now exacerbated by the increasing use of smart phones, tablets and social network type solutions hiding even more knowledge from us poor, hard pressed IT foot soldiers during data migrations.
In the past, it has to be said, from an enterprise application migration perspective, our project mission statement was often to wipe out this alternative processing provision, but this attitude is changing.
First though let’s take a step back and have a look at what types of data migration there are out there. If I use a categorisation Philip Howard of Bloor Research proposed at DMM3, there are at least five types of data migration.
The first he called ’Disk to Disk’ migration and here he included migrations within data-centres as well as local movements. He suggested that this kind of migration faced its largest challenges in determining what you have that is worth keeping. But here, obviously, the maintenance of links is paramount.
The second category Phillip identified was the database to database migration - maybe re-platforming and application on SQL Server from DB2. The appropriate tools here may include those from my friends at Rever but link preservation is not often at issue.
Thirdly we have what he termed content migration. This covers the movement of unstructured data to new repositories and here we definitely are in the realm of the need to manage url and embedded link pointers.
Fourthly there is spreadsheet management and governance. This is not an area where I have much experience but, from what I understand, it is common in those environments (usually financial) that have given themselves over to rule by Excel to try to put some order on things, possibly by moving their key spreadsheets to a central repository. And here, again, link preservation is essential.
Finally we have my corner of the sandpit - what I call enterprise application migration or, as Phillip terms it, ’application centric migration’. This is the populating of new enterprise applications (think SAP) with data at go live. (It is also the populating of existing applications with data from other applications in mergers, de-mergers or acquisitions but let’s not get too precise).
However, as I pointed out in the keynote presentation at DMM5, even within the couple of years between Phillips presentation and mine, these clear boundaries are beginning to blur. In the last 12 months I have been involved with more than one migration where the target architecture included an enterprise application (Microsoft Dynamics) but also linking to a document repository (SharePoint). So it is increasingly common now to have a mix of type 3 and type 5. As I also pointed out in my DMM5 spiel - with the advent of cloud computing as a real enterprise option, the distinct domains of migration types 1 and 3 are also beginning to overlap.
All of which brings us back to LinkFixer Advanced, the premier product of LinkTec, Ed Clark the COO of whom sparked this train of thought. LinkFixer Advanced is designed (as its name suggests) to find and fix broken links, in batches of thousands at a time. So obviously a move of unstructured data to SharePoint would be right up its street. It has a pre-emptive option where it will catalogue links prior to a move, making re-linking easier and automatic. Talking with Ed was illuminating because as a company they had thought their product would find most take up in the web design and maintenance sphere but instead they are finding that it is in data migration scenarios where they are experiencing greatest traction. All of which I guess illustrated my thesis that there is a convergence going on here. We have to accept that what we could once ignore (or as I indicated, actively try to suppress) is increasingly becoming part of our project scope. So we all need to be aware of the existence of tools like LinkFixer Advanced so that we can be ready to meet the challenge when we hit it.
The presentations I referred to can be found on the Data Migration Matters website - datamigrationmatters.co.uk.
Twitter - @johnymorris
About the author
John Morris has over 20 years experience in IT as a programmer, business analyst, project manager and data architect. He has spent the last 10 years working exclusively on data migration and system integration projects. John is the author of Practical Data Migration.